/blockscope

This package allows for variables to be block scoped, a common feature of many languages but not Python's

Primary LanguagePythonMIT LicenseMIT

Block Local Scope

This simple package allows for variables to be block scoped, which is a common feature of many languages but is missing in Python.

Installation & Basics

  $ pip install blockscope

The package exposes a single class Local. An instance of this class is meant to be used with the with block.

For the most basic use cases, the keyword arguments to the Local constructor declare and initialize "variables" that become attributes of the Local instance:

    from blockscope import Local

    with Local(x=1, y=2) as local:
        print( local.x, local.y )  # prints: 1 2
        local.z = 3                # declare new variable
        print( local.z )           # prints: 3

    # when `with` exits: local's x, y, and z
    # are cleaned up automatically

Note that you can add as many new variables as you'd like inside the with block, such as the local.z variable shown above.

When the with block exits, all attributes of Local instance are released.

Tuple Unpacking

One of the more useful features is tuple (or any iterable) unpacking, which allows you to quickly and easily assign multiple values at once to separate variables. For example:

    def foo():
        return 1, 2, 3

    with Local('x, _, z', foo()) as local:
        print( local.x, local.z )  # prints: 1 3

Note the placeholder _ to ignore the second element of the tuple. This is in line with Python's regular syntax for tuple unpacking, such as x, _, z = foo()

You can use *, ?, and ~ modifiers to fine-tune the unpacking process. Consider this example:

    def bar():
        return 4, 5

    with Local('x, _?, y? , z~ ,_*', bar()) as local:
        # x   Sets local.x to the first element of the tuple (4).
        # _?  ignores the second element of the tuple (5) if present.
        #     If _ is by itself (no ?) and the element is not present
        #     an AttributeError with a helpful message is raised.
        # y?  Sets local.y to the 3rd element of the tuple if present,
        #     In this case, there's no 3rd element and local.y will
        #     not exist.
        # z~  local.z would be set if present or set to None otherwise
        #     In this case, local.z will be None.
        # _*  Ignores the rest of the tuple elements, and is always used
        #     in the last position with a placeholder _ or by itself.
        print( local.x, local.z) # prints: 4 None

You can chain multiple unpackings and declarations together, allowing you to extract and assign values from multiple tuples or iterables in a single line of code.

    with Local('x,y,*', foo(), '_,z' = bar(), i=6, j=7) as local:
        print( local.x, local.y, local.z )  # prints: 1 2 5
        print(i, j) # prints: 6 7

Dictionary and List Unpacking

Dictionaries {str:any} and lists of tuples [(str, any)] are unpacked automatically if not assigned to a keyword argument:

    d = {'x':1, 'y':2, 'z':3}
    l = [('a': 4), ('b': 5), ('c': 6)]

    with Local(d, l) as local:
        print( local.x, local.y, local.z ) # prints: 1 2 3
        print( local.a, local.b, local.c ) # prints: 4 5 6